Measurements of whole cores taken at Site 898 included magnetic susceptibility, Gamma-Ray Attenuation Porosity Evaluator (GRAPE) bulk densities, P-wave-logger (PWL) velocities, and thermal conductivity. Reliable velocities were obtained in unlithified sediments using the Digital Sound Velocimeter (DSV) on split cores and, within the more consolidated units, using the Hamilton Frame Velocimeter. Electrical resistivity and undrained shear strength were measured on split cores. Index properties were calculated from wet and dry masses and wet and dry volumes. Physical-properties data generally show uniform downhole trends for the 342 m of Hole 898 A. Only one APC-core was taken at Hole 898B. Data from this core are discussed together with the data from Hole 898A.
Index properties were determined using gravimetric methods (Table 10; Fig. 25). Densities also were determined, with an uncertainty of ±0.02 g/cm3. The accuracy of the porosity measurements has been estimated as 2%. Bulk and grain densities and porosity distributions indicate smooth downhole trends. Gravimetrically determined bulk densities increase from about 1.7 g/cm3 near the seafloor to 2.0 g/cm3 at a depth of 339 mbsf. Porosities decrease from about 65% at the seafloor to 49% at the base of the hole. Grain densities are nearly constant with depth and have a mean value of 2.78 g/cm3.
The excellent APC-core recovery in the upper 132 m of Hole 898 A preserved thick sand units, which provided us an opportunity to sample systematically both clay- and sand-rich lithologies and to compare their respective index properties. However, sand units near the top of the sediment section were soupy and had been disturbed by drilling; therefore, the calculated porosities must be viewed with caution. Porosities obtained from discrete sand samples are indicated in Figure 25 as open diamonds. Porosities in sands generally are lower than those for clay-rich units (Hamilton, 1976), and range from about 50% to 55% in the upper 150 m of the hole.
Bulk densities were estimated also from whole-core GRAPE measurements taken on all sections recovered from Holes 898A and 898B (see "Explanatory Notes" chapter, this volume). The maximum GRAPE densities give the best estimate for true bulk density of the sediment (Boyce, 1973; Gealy, 1971). The visually estimated maximum densities are indicated by the line superimposed on the Gravimetric Bulk Density and GRAPE Bulk Density graphs in Figure 25. The line shows a slight net decrease in bulk density with depth. This contrasts with the gravimetrically determined bulk density, which shows an increase down the hole from 1.7 g/cm3 near the seafloor to 2.0 g/cm3 at 339 mbsf (Fig. 25). The GRAPE density appears to be high relative to the average gravimetric density near the top of the hole and low near the base of the hole. This discrepancy is probably related, in part, to incompletely filled liners in the deeper XCB cores.
Electrical resistivity was measured at intervals of 0.5 to 0.75 m in all split cores down to 340 mbsf. Formation factors calculated from the resistivity data (see "Explanatory Notes" chapter, this volume) show a general increase with depth (Fig. 25). In the upper 230 m of sediment, the formation factor ranges from 4 to 6, defining a fairly linear downhole trend. Below 230 mbsf, the formation factor is consistently higher and exhibits a large scatter between 5 and 15. This scatter can be attributed to the presence of less porous nannofossil clays (see "Lithostratigraphy" section, this chapter).
Using the vane shear apparatus, undrained shear strength was measured in cores from the upper 260 m of Hole 898A at a frequency of one measurement per core (Fig. 26). Shear strengths ranged from about 6 kPa at a depth of 2 mbsf (Core 149-898A- 1H) to a maximum of 122 kPa at 162 mbsf (Core 149-898A-18X). The scatter of the shear strength data increases below 100 mbsf, most likely as a consequence of sampling different lithologies.
Discrete acoustic velocity was measured in each core recovered from Hole 898A (Table 11). No velocity measurements were performed in the single core from Hole 898B. The DSV was used for Cores 149-898A-1H to -10H to provide compressional-wave velocities in weakly consolidated sediments shallower than 95 mbsf. The Hamilton Frame Velocimeter was used to measure velocities in samples taken from Cores 149-898A-11H to -36X. Compressional-wave velocities in Cores 149-898A-11H to -36X were measured in three mutually orthogonal directions to assess the degree of acoustic anisotropy (see "Explanatory Notes" section, this volume). Repeated measurements of selected samples and calibration standards suggest an accuracy of 2% to 3% for the velocity measurements.
Discrete acoustic velocity measurements show a general increase with depth, from about 1500 m/s at the seafloor to about 1700 m/s at 340 mbsf (Fig. 27). The increase downhole in the vertical component of velocity shows a reasonably linear trend, with a slope of 0.58 s-1. The correlation coefficient for the fit is 0.78, which reflects the large scatter of the downhole velocity measurements. The horizontal components of velocity show similar downhole variations. The relatively high velocities at approximately 100 and 170 mbsf were observed in all three directions. Acoustic anisotropy at these depths is approximately 8%. Anisotropy throughout the remainder of the hole is generally less than 5%, which is less than the estimated error.
Compressional velocity also was measured with the PWL tool in APC Cores 149-898A-1H to -14H. Cores below 149-898A-14H, recovered with the XCB bit, were not of the quality for reliable PWL measurements. The PWL data are shown in Figure 28, along with the linear velocity gradient derived from the discrete velocity measurements. Data having poor signal strength are not shown (see "Explanatory Notes" chapter, this volume). Above 30 to 40 mbsf, the highest PWL velocities agree well with the discrete velocity measurements. Lower PWL velocities in this interval represent measurements taken in sections of core that had been disturbed by drilling or that contained visible gas bubbles and, therefore, are not considered representative of the sediment velocities. PWL velocities below 40 mbsf are systematically lower than the discrete velocity measurements. Possible reasons for this discrepancy include increased gas content in the core, which would reduce PWL velocities, and biased sampling for the discrete velocity measurements. Sample bias occurred because velocity measurements were taken in sediment that was sufficiently coherent to withstand either the insertion of the DSV transducers or removal of the sample for measurement in the Hamilton Frame Velocimeter. This resulted in preferential sampling of clay and ooze. PWL measurements were taken before the core liner was split and, therefore, also included incoherent sand and silt.
Magnetic susceptibility was measured at intervals of 3 to 5 cm in all cores collected at Site 898. The results are discussed in the "Paleomagnetism" section (this chapter).
Thermal conductivity was measured in alternate sections of all cores from Site 898. The mean error was estimated as ±0.2 W/(m·K). Thermal conductivity values increase slightly with depth about a linear trend of 1.1 W/(m·K) at the seafloor to 1.5 W/(m·K) at 340 mbsf (Fig. 29; Table 12). This increase most likely is the result of reduced porosity arising from compaction. From 0 to 50 mbsf, most measurements lie near 1.25 W/(m·K). Between 50 and 140 mbsf, the values exhibit a large scatter about 1.4 W/(m·K). At 140 mbsf, thermal conductivity abruptly decreases to 1.2 W/(m·K). Below this depth, a consistent increase of conductivity with depth to about 1.6 W/(m·K) at 280 mbsf can be observed. The measurements are more scattered again below 280 mbsf.